Skip to Content

Inside China’s Genome Factory

Sequencing a complete human genome may soon cost less than an iPhone. Will BGI-Shenzhen decode yours?
February 11, 2013

When he was 17 years old, Zhao Bowen dropped out of Beijing’s most prestigious high school. Like many restless young people in China, he headed south to Shenzhen, the country’s factory capital, for a job. As a teenage science prodigy, however, he wasn’t bound for an assembly-line floor; instead, he was on his way to the world’s largest production center for DNA data. Now, a few years later, in a retrofitted shoe factory that is the headquarters of BGI-Shenzhen, the 21-year-old is orchestrating an effort to decipher the genetic makeup of some 2,000 people—more than 12 trillion DNA bases in all.

BGI-Shenzhen, once known as the Beijing Genomics Institute, has burst from relative obscurity to become the world’s most prolific sequencer of human, plant, and animal DNA. In 2010, with the aid of a $1.58 billion line of credit from China Development Bank, BGI purchased 128 state-of-the-art DNA sequencing machines for about $500,000 apiece. It now owns 156 sequencers from several manufacturers and accounts for some 10 to 20 percent of all DNA data produced globally. So far, it claims to have completely sequenced some 50,000 human genomes—far more than any other group.

BGI’s sheer size has already put Chinese gene research on the map. Those same economies of scale could also become an advantage as comprehensive gene readouts become part of everyday medicine. The cost of DNA sequencing is falling fast. In a few years, it’s likely that millions of people will want to know what their genes predict about their health. BGI might be the one to tell them.

The institute hasn’t only initiated a series of grandly conceived science projects. (In January, it announced it had determined the DNA sequence of not one but 90 varieties of chickpeas.) It’s also pioneered a research-for-hire business to decode human genomes in bulk, taking orders from the world’s top drug companies and universities. Last year, BGI even started to install satellite labs inside foreign research centers and staff them with Chinese technicians.

BGI’s rise is regarded with curiosity and some trepidation, not just because of the organization’s size but also because of its opportunistic business approach (it has a center for pig cloning, dabbles in stem-cell research, and runs a diagnostics lab). The institute employs 4,000 people, as many as a midsize university—1,000 in its bioinformatics division alone. Like Zhao, most are young—the average age is 27—and some sleep in company dormitories. The average salary is $1,500 a month.

Ten years ago, the international Human Genome Project was finishing up the first copy of the human genetic code at a cost of $3 billion. Thanks to a series of clever innovations, the cost to read out the DNA in a person’s genome has since fallen to just a few thousand dollars. Yet that has only created new challenges: how to store, analyze, and make sense of the data. According to BGI, its machines generate six terabytes of data each day.

Zhang Yong, 33, a BGI senior researcher, predicts that within the next decade the cost of sequencing a human genome will fall to just $200 or $300 and BGI will become a force in assembling a global “bio-Google”—it will help “organize all the world’s biological information and make it universally accessible and useful.”

BGI quotes prices as low as $3,000 to sequence a person’s DNA.

Some outsiders, however, question whether BGI is anything more than biology’s version of Foxconn, the giant assembler of iPads and other gadgets designed elsewhere, whose largest factory, employing some 240,000 workers, is also in Shenzhen. While BGI has done important science—a recent paper on sequencing the bacteria in the human gut made the pages of Nature—it’s seen more as a mass producer of data than as an instigator of original research that can explain what the results mean.

“BGI has scaled up very impressively,” says Eric Lander, director of the Broad Institute in Cambridge, Massachusetts, which operates the largest academic DNA sequencing center in the United States. “But I think that absolute scale is much less important than what you do with it.”

“Don’t Worry, Be Happy”

BGI’s president, Wang Jian, 59, cofounded the company with Yang Huanming, 61, in 1999. They managed to persuade the leaders of the Human Genome Project, then in full swing, to let them handle 1 percent of the work, making China the only developing nation to play a major role. In 2002, BGI turned heads by publishing the complete sequence of the rice plant in Science. Research in the national interest has been a BGI mainstay: it decoded the DNA of the giant panda, and it discovered the genetic mutation that makes Tibetans so well suited to life at high altitudes. Outside work hours, Wang is known for having scaled Mount Everest in 2010. (“It’s a national park—so what? Not a big deal,” he says.)

A quirky, informal logic governs BGI. That has made it a puzzle to observers; it’s very different from hierarchical Chinese institutions, where credentials and connections can matter most. Deputy director Xu Xun, 29, who oversees the 1,000-strong bioinformatics group, says it’s why BGI attracts so many talented young people. “You get [to play a role in] a lot of exciting things here,” he says. In 2010, Nature cited BGI’s model in an editorial questioning whether scientists really need PhDs. Xu himself came to BGI after abandoning his PhD studies. He’s what’s known admiringly around the company as a “leaver”—impatient with school and eager for real-world experience.

Blood samples are fed into automated DNA sequencing machines.

In Wang’s cubicle, which is in the middle of a long bay of identical blue cubicles backed by windows overlooking a mountainside construction site, is a hand-signed letter from Bill Gates announcing a partnership on agricultural genomics that BGI and the Gates Foundation agreed to last fall. “He loves the dropouts,” Wang says of the Microsoft chairman. He grins and then sings a few bars from Bobby McFerrin’s “Don’t Worry, Be Happy,” adding, “I love that song.”

BGI bills itself as China’s first “citizen-managed” research institute. With strong political support from Beijing, it became part of the vaunted Chinese Academy of Sciences in 2003. But it was an uneasy fit. The conservative academy limits the size of its institutes, yet BGI was on a hiring spree, and it was thumbing its nose at university credentials, too. In 2007, the government of Shenzhen offered BGI $12.8 million to move to the port city and become an independent institute.

Today, Wang says, only about 10 percent of BGI’s revenue comes from government projects—and that’s largely from local municipalities, not from Beijing. The rest is a mix of grants, some anonymous donations, and fees from clients, including as little as $3,000 to $4,000 to sequence a human genome.

And even though it’s a nonprofit, BGI operates several businesses. That’s left Westerners guessing at the true nature of the Chinese institute. Wang says they shouldn’t worry. “We like science; we need money. We put the two things together,” he says. “I use my left hand to make money and my right hand to do basic research.” At a recent biotechnology conference in Shenzhen, cosponsored by BGI, Wang gave the opening presentation. One of his slides read: “World-Class Science = World-Class Business.”

Thousands of Genomes

In its scientific work, BGI often acts as the enabler of other people’s ideas. That is the case in a major project conceived by Steve Hsu, vice president for research at Michigan State University, to search for genes that influence intelligence. Under the guidance of Zhao Bowen, BGI is now sequencing the DNA of more than 2,000 people—mostly Americans—who have IQ scores of at least 160, or four standard deviations above the mean.

The DNA comes primarily from a collection of blood ­samples amassed by Robert Plomin, a psychologist at King’s College, London. The plan, to compare the genomes of geniuses and people of ordinary intelligence, is scientifically risky (it’s likely that thousands of genes are involved) and somewhat controversial. For those reasons it would be very hard to find the $15 or $20 million needed to carry out the project in the West. “Maybe it will work, maybe it won’t,” Plomin says. “But BGI is doing it basically for free.”

From Plomin’s perspective, BGI is so large that it appears to have more DNA sequencing capacity than it knows what to do with. It has “all those machines and people that have to be fed” with projects, he says. The IQ study isn’t the only mega-project under way. With a U.S. nonprofit, Autism Speaks, BGI is being paid to sequence the DNA of up to 10,000 people from families with autistic children. For researchers in Denmark, BGI is decoding the genomes of 3,000 obese people and 3,000 lean ones.

Beyond basic science, BGI has begun positioning itself as the engine of what’s expected to be a boom in the medical use of genome scans. In 2011, for instance, it agreed to install a DNA analysis center inside the Children’s Hospital of Philadelphia, a leading pediatric hospital. Ten bioinformatics experts were flown in from Shenzhen on temporary visas to create the center, which opened six months later with five sequencing machines.

As the technology enters clinical use, the number of genomes sequenced in their entirety could catapult into the millions per year. That is what both the Philadelphia hospital and BGI are preparing for. “They have the expertise, instruments, and economies of scale,” says Robert Doms, pathologist-in-chief of the children’s hospital. He says it will pay BGI a fee for each genome it sequences, and will offer the service to parents of young patients with undiagnosed diseases. The hospital will also be developing new types of genetic tests, an area where the Chinese agree they have much to learn. Although BGI operates a genetic testing center in China, the degree of regulation seen in the United States is new to its researchers. “It’s a whole additional level of rigor,” says Doms.

BGI is also proving it can be nimble in seeking business opportunities. Last fall, it made a surprise bid to purchase a faltering U.S. company, Complete Genomics of Mountain View, California. The company operates a complex technology for sequencing human DNA; in 2012 it accounted for perhaps 10 percent of all human DNA data generated globally. But it was losing money.

BGI’s bid to buy the company, for the fire-sale price of $118 million, has stirred competitive worries in the U.S. The main supplier of DNA sequencing instruments, Illumina, tried to break up the deal with a counter-bid and appealed to Washington to block the takeover. Letting BGI snap up the company would be equivalent to selling China the “formula for Coke,” said Illumina’s CEO, Jay Flatley. Flatley cautioned that the Chinese, until now dependent on U.S. machinery, could dominate next-generation technology—and that they could even somehow make “nefarious” use of American DNA data flowing through their computer servers by the terabyte. U.S. regulators have dismissed the national security concerns, and approval of the deal is pending.

BGI’s leaders know that resistance to its expansion is real, but they shrug off the concerns. The joke in the cubicles is that if BGI were truly a tool of Beijing, it would probably have nicer office space. More matter-of-factly, Xu, the head of the bioinformatics team, says of the acquisition: “This is a good technology. The company is bankrupt; we think it’s a good chance to do something.”

Wang, the Everest climber, is still frequently asked to explain BGI’s strategy and its intentions. He says to think of a wandering migrant worker—looking for opportunity and occasionally irritating the authorities. That is what BGI is like. But its only core mission is to do work that will be socially useful, he says: its strategy is to “do good.”

Keep Reading

Most Popular

It’s time to retire the term “user”

The proliferation of AI means we need a new word.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Sam Altman says helpful agents are poised to become AI’s killer function

Open AI’s CEO says we won’t need new hardware or lots more training data to get there.

An AI startup made a hyperrealistic deepfake of me that’s so good it’s scary

Synthesia's new technology is impressive but raises big questions about a world where we increasingly can’t tell what’s real.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.